The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses curl to interact with the brand.dev API and to download logo assets into the local file system.
[EXTERNAL_DOWNLOADS]: Fetches data and image files from the external brand.dev domain. This is the primary function of the skill and uses standard HTTPS requests.
[PROMPT_INJECTION]: There is a potential for indirect prompt injection because the skill processes untrusted data (brand names and descriptions) from an external API.
Ingestion points: Data enters the agent context from the api.brand.dev response as described in SKILL.md.
Boundary markers: No specific delimiters are defined for the fetched brand description.
Capability inventory: The skill has file system write access (mkdir, curl -o) and network access (curl).
Sanitization: The skill includes instructions to sanitize the domain input and format the brand slug (lowercase, hyphens instead of spaces) before using them in commands.

brand-research